Transaction Recurrence Prediction via Supervised Learning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for identifying recurring payment relationships between merchants and customers rely on manual rule sets and customer feedback, leading to inaccurate determinations due to subjective interpretation and incomplete data.

Innovation Solution

A system that aggregates transaction data to predict recurrence periods using supervised learning models, determining probability scores for recurring relationships without relying on customer feedback, by analyzing transaction dates and splitting data into training and testing subsets to identify recurring patterns.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual rule sets and customer feedback are used to identify recurring relationships, then the system is simple to implement, but the accuracy of identification deteriorates due to subjective interpretation and incomplete data

Engineering Contradiction:

Improveaccuracy of identificationVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces manual rule-based systems with machine learning models that automatically analyze transaction patterns. Supervised learning algorithms process transaction data to identify recurring relationships without relying on manual rule creation or customer feedback, thereby improving accuracy while managing complexity through automated computational processes.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by allowing the machine learning model to autonomously learn from transaction data and make determinations about recurring relationships without human intervention. The model continuously improves through training on historical data, performing the identification task independently rather than relying on manual rules or customer input.

Inventive Principle:

Principle #25Self-service

2Reliability

If customer feedback is solicited to determine recurring relationships, then the system can incorporate customer perspective, but the process becomes slower and less reliable due to customer inaction and subjective interpretation

Engineering Contradiction:

Improvereliability of determinationVSAvoidtime for determination

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary action by pre-processing and analyzing transaction data continuously in the background. The machine learning model is trained on historical transaction patterns beforehand, so when a determination is needed, the system can quickly query the trained model rather than waiting for customer feedback or performing complex analysis at the moment of need.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces the mechanical process of soliciting and processing customer feedback with an automated machine learning system that analyzes transaction data objectively. This substitution eliminates delays caused by customer inaction and removes subjective interpretation, providing reliable determinations instantly based on learned patterns from historical data.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Ease of manufacture

If simple manual rule sets are used with boundaries on mean and standard deviation, then the system is easy to implement, but the accuracy of recurring relationship identification deteriorates

Engineering Contradiction:

Improveease of implementationVSAvoidaccuracy of identification

Core Design Contradiction:

Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent replaces simple manual rule sets with sophisticated machine learning models that can capture complex, non-linear patterns in transaction data. These models automatically learn optimal thresholds and relationships from training data, eliminating the need for manual rule creation while significantly improving identification accuracy through pattern recognition capabilities that far exceed simple statistical boundaries.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11144935B2Technique to aggregate merchant level information for use in a supervised learning model to detect recurring trends in consumer transactions

Publication Date: 2021.10.12 CAPITAL ONE SERVICES LLC

AI summary

A method is disclosed, comprising: aggregating a plurality of sets of transactions, each set of transactions comprising transactions related to an account-merchant pairing; determining variables characterizing a recurrence period based on transaction dates of the transactions in the each set; predicting the recurrence period for a transaction related to the account-merchant pairing for each customer of the plurality of customers; aggregating another set of transactions between the plurality of customers and the merchant; evaluating a distribution of the recurrence period for each customer within range of a distant point; and based on the evaluation of the distribution of the recurrence period for each customer, generating a probability of the merchant having a recurrent transaction with the customer. The account-merchant pairing may comprise a customer account and a merchant of a plurality of customers and merchants. Another set of transactions may comprise transactions in the plurality of sets of transactions.